Skip to content
Research Article Open access CC BY 4.0

Diagnosis of Melanoma Based on the Sparse Auto-Encoder for Feature Extraction

Nadia Smaoui Zghal, Marwa Zaabi, Houda Derbel

Annual Research & Review in Biology · pp. 220–230 · Published 31 Dec 2020

10.9734/arrb/2020/v35i1230327

Abstract

Aims: Skin cancer is a fairly critical disease all over the world and especially in Western countries and America. However, if it is perceived and treated early, it is quite often curable. The main risk factors for melanoma are exposure to UV rays, the presence of many moles, and heredity. For this reason, this work focuses on the issue of automatic diagnosis of melanoma. The aim is to extract significant features from pixels of the images based on an unsupervised deep learning technique which is the sparse autoencoder method. Methodology: A preprocessing phase is required to remove the artifacts and enhance the contrast of the images before proceeding with the feature extraction. Once the characteristics are extracted automatically, the support vector machine classifier and the k-nearest neighbors are applied for the classification phase. The objective is to differentiate between 3 categories: melanoma, suspected case, and non-melanoma. Finally, the PH2 database is used to test the proposed approaches (200 images are presented in this dataset: 80 atypical nevi, 80 common nevi, and 40 melanoma). Results: The obtained results in terms of specificity, accuracy, and sensitivity present noticeable performances with the support vector machine classifier (achieved 94 % overall accuracy) and the k-nearest neighbors (92 %). Conclusion: This study's experimental findings showed that the best performance was obtained by the approach based on a deep sparse autoencoder combined with support vector machine.

Melanoma sparse autoencoder support vector machine k-nearest neighbors.

References (35)

  1. 1 Deep Learning in Medical Image Analysis [DOI]
  2. 2 An Introduction to Support Vector Machines and Other Kernel‐based Learning Methods [DOI]
  3. 3 Skin lesion image segmentation using Delaunay Triangulation for melanoma detection [DOI]
  4. 4 The melanoma skin cancer detection and classification using support vector machine [DOI]
  5. 5 Interactive atlas of dermoscopy
  6. 6 Automatic Detection of Melanoma with Yolo Deep Convolutional Neural Networks [DOI]
  7. 7 A Method for Melanoma Skin Cancer Detection Using Dermoscopy Images [DOI]
  8. 8 Automatic melanoma detection via multi-scale lesion-biased representation and joint reverse classification [DOI]
  9. 9 A Color-Based Approach for Melanoma Skin Cancer Detection [DOI]
  10. 10 Advanced earlier melanoma detection algorithm using colour correlogram [DOI]
  11. 11 Skin lesion analysis system for melanoma detection with an effective hair segmentation method [DOI]
  12. 12 Place and role of intelligent systems in computer science [DOI]
  13. 13 Lesion Attributes Segmentation for Melanoma Detection with Multi-Task U-Net [DOI]
  14. 14 Convolutional Neural Network Algorithm with Parameterized Activation Function for Melanoma Classification [DOI]
  15. 15 Deep Convolutional Neural Network for Melanoma Image Classification [DOI]
  16. 16 Lesion Segmentation and Automated Melanoma Detection using Deep Convolutional Neural Networks and XGBoost [DOI]
  17. 17 Detection of skin cancer ”Melanoma” through computer vision [DOI]
  18. 18 Automatic diagnosis of melanoma using linear and nonlinear features from digital image [DOI]
  19. 19 Melanoma Detection Using Delaunay Triangulation [DOI]
  20. 20 Automatic diagnosis of melanoma from dermoscopic image using real-time object detection [DOI]
  21. 21 Detection of melanoma using distinct features [DOI]
  22. 22 Melanoma Skin Cancer Detection Based on ABCD Rule [DOI]
  23. 23 Feature selection using sequential backward method in melanoma recognition [DOI]
  24. 24 Two methodologies for identification of stages and different types of melanoma detection [DOI]
  25. 25 MelanomaNet: An Effective Network for Melanoma Detection [DOI]
  26. 26 Hill Climbing Optimization and Fuzzy C-Means Clustering for Melanoma Skin Cancer Identification and Segmentation [DOI]
  27. 27 A Hybrid Diagnosis System for Malignant Melanoma Detection in Dermoscopic Images [DOI]
  28. 28 Segmenting melanoma Lesion using Single Shot Detector (SSD) and Level Set Segmentation Technique [DOI]
  29. 29 DermaDL: Advanced Convolutional Neural Networks for Automated Melanoma Detection [DOI]
  30. 30 Automatic Detection and Segmentation of Melanoma using Fuzzy C-Means [DOI]
  31. 31 Deep Attention Network for Melanoma Detection Improved by Color Constancy [DOI]
  32. 32 Comparative Study of Two Classification Methods for the Detection of Alzheimer's Disease [DOI]
  33. 33 Computer Aided Early Detection and Classification of Malignant Melanoma [DOI]
  34. 34 26th Annual International Conference of the IEEE EMBS
  35. 35 International Journal of Machine Learning and Computing [DOI]

Cited by 2

Evaluation of denoising techniques to remove speckle and Gaussian noise from dermoscopy images

Evgin Goceri · Computers in Biology and Medicine · 2023

Showing 1 of 2 known citations — external sources report more than can currently be individually listed.

Related research

Article metrics

Real usage data collected on this platform.

63

Page views

0

PDF downloads

10

Outbound clicks

2

Citations

Views over time

Views by country

Approximate, from request IP at view time — not citizenship or institution. Countries with fewer than 5 views are grouped as "Other".

Traffic sources

Referring site, by host.

Views and downloads exclude known bots/crawlers. Citations combines this platform's own DOI-resolved index with each external source's own reported total — see Cited by above for individually listed citing works. Last refreshed 0 seconds ago.